Hello @Borda! Thanks for updating the PR.

• In the file skimage/exposure/exposure.py, following are the PEP8 issues :

Line 273:80: E501 line too long (91 > 79 characters)
Line 279:80: E501 line too long (91 > 79 characters)

• In the file skimage/feature/blob.py, following are the PEP8 issues :

Line 148:80: E501 line too long (100 > 79 characters)

• In the file skimage/feature/corner.py, following are the PEP8 issues :

Line 256:80: E501 line too long (84 > 79 characters)
Line 293:80: E501 line too long (82 > 79 characters)

• In the file skimage/feature/orb.py, following are the PEP8 issues :

Line 102:80: E501 line too long (86 > 79 characters)
Line 108:80: E501 line too long (86 > 79 characters)

• In the file skimage/measure/block.py, following are the PEP8 issues :

Line 41:80: E501 line too long (97 > 79 characters)

• In the file skimage/measure/fit.py, following are the PEP8 issues :

Line 383:80: E501 line too long (81 > 79 characters)

• In the file skimage/measure/profile.py, following are the PEP8 issues :

Line 48:80: E501 line too long (83 > 79 characters)

• In the file skimage/transform/integral.py, following are the PEP8 issues :

Line 71:80: E501 line too long (91 > 79 characters)

Comment last updated on January 16, 2018 at 22:06 Hours UTC

well, it seems that what is passing in PIP_FLAGS="--pre" doe not pass in WITH_PYAMG=1 MINIMUM_REQUIREMENTS=1 WITH_QT=1

@Borda you have lots of unrelated changes in this PR as well. If you Enable access for the maintainers, I could help with this issue.

@soupault I have check the option Allow edits from maintainers. in this PR or do you mean something else?

@Borda yes, thanks! I've rebased the branch onto upstream master.

However, I'm not able to fix your/origin master branch (the granted permissions does not allow this). You'd need to do this by yourself. See https://stackoverflow.com/a/42332860/1759565 for the instructions.

I did as the post recommends and I hope that it is correct now...

I think the 1st commit (Merge remote-tracking ...) is unwanted. Otherwise, should be ok.

Thanks for all your hard work @Borda and @soupault

Did you consider using np.set_printoptions(legacy='1.13') ? (see for example https://github.com/nilearn/nilearn/pull/1592/files)

Update: OK, it was mentioned in #2935

@soupault I see, so hope that now it is good...
@emmanuelle well, that could be also an option... so which way should we go?

@Borda I'm closing this PR as we decided to move on with np.set_printoptions(legacy='1.13'). All deprecations related to numpy=1.14 are to be addressed in #2935.

Thank you for your attempt on fixing this, and you are welcome to contribute to the discussion in #2935!